Skip to main content

What Is a Demand Charge? Load Factor and Your Real Rate per kWh

Two businesses on one tariff can pay 22.27 cents and 14.57 cents a kilowatt-hour, decided by nothing but how evenly they drew power.

By Mohamed Zakrya

Updated · 9 min read

Share
A bill that charges for how much and how hard What is a demand charge One bill charges for how much you used. The other charges for how hard you pulled. 1 · HOW MUCH Kilowatt-hours, summed across the whole month. 3,000 × $0.12 $360.00 2 · HOW HARD Kilowatts, read from the worst 15 minutes. 10 kW × $15 $150.00 3 · LOAD FACTOR kWh ÷ (kW × 730 h) 41.1% Average power against the peak. Not on the bill. 4 · THE REAL RATE $510 ÷ 3,000 kWh $0.1700 The tariff says 12 cents. Demand adds five. AND LOAD FACTOR IS WHAT SETS THAT ADDER One tariff, two very different prices At a 20 percent load factor the all-in rate is 22.27 cents a kilowatt-hour. At 80 percent it is 14.57. Nothing about the rate card differs between them. 52.9 percent apart, decided by shape alone Which lever pays A point off usage is worth $3.60 here; a point off the peak is worth $1.50. The ratio is just $360 ÷ $150, so the bigger charge always wins per point. they break even at a 50 percent demand share Both rates here are illustrative inputs, not market figures. Real tariffs add demand ratchets, seasonal rates, time-of-day demand windows, coincident-peak charges and power-factor penalties, none of which this two-part model carries. The tariff sheet and the interval data are the source of truth.
Two businesses on one tariff can pay 22.27 cents and 14.57 cents a kilowatt-hour, decided by nothing but how evenly they drew power.

A commercial electricity bill can charge for two different things. The energy charge measures how much electricity accumulated across the billing period, while the demand charge measures how hard the customer drew power at its most demanding moment.

Those quantities use different units. Energy is billed in kilowatt-hours, while demand is billed in kilowatts, because it describes a rate of power rather than an accumulated amount.

Demand is usually read from the highest 15-minute interval recorded during the billing period, so one difficult quarter-hour can shape the bill even when monthly consumption looks ordinary. That is why two businesses can consume identical amounts of electricity and still receive very different bills: their kilowatt-hours match, and the shape of their demand does not.

One bill, two different charges

The default inputs in the demand charge calculator are a 10-kilowatt peak, 3,000 kilowatt-hours of monthly consumption, a demand rate of $15 per kilowatt, and an energy rate of $0.12 per kilowatt-hour.

Both rates are illustrative inputs rather than market figures or any particular tariff. The tool page describes the $15 per kilowatt as an illustrative middle of a range the National Renewable Energy Laboratory found spanning under $5 to over $50 across more than 10,000 United States commercial and industrial tariffs, and the 12 cents is illustrative in the same way.

Demand charge = 10 kW × $15 per kW = $150.00
Energy charge = 3,000 kWh × $0.12 per kWh = $360.00
Total bill = $150.00 + $360.00 = $510.00
Demand share = $150.00 ÷ $510.00 × 100 = 29.4 percent

The customer pays $360.00 for the energy accumulated through the month and another $150.00 for reaching a 10-kilowatt peak, so demand accounts for 29.4 percent of this two-part bill.

Kilowatts and kilowatt-hours are related but not interchangeable. A kilowatt states how fast power is being drawn at a moment or across a measurement interval, while a kilowatt-hour records how much energy accumulated as that rate continued through time.

The infrastructure serving a customer has to accommodate that customer's hardest interval rather than their monthly average. The demand charge prices the capacity requirement and the energy charge prices the electricity delivered, which is the whole reason a tariff separates them.

This structure is normally associated with commercial and industrial accounts. Residential customers should not be assumed to pay demand charges as a rule, although some utilities have introduced residential demand rates in particular service territories.

Dividing the complete bill by its consumption puts both charges over the same denominator and produces an effective all-in rate.

Effective all-in rate = $510.00 ÷ 3,000 kWh = $0.1700 per kWh

The posted energy rate is 12 cents a kilowatt-hour, but this bill works out at 17 cents once demand is included. Load factor explains the five-cent difference.

Load factor connects the two halves

Load factor compares average power against peak power. A high load factor means demand stayed relatively even; a low one means the peak stood far above the level sustained through the rest of the month.

The calculator does not compute it, because it needs the length of the billing period and the engine has no such input. For a monthly figure this guide divides a 365-day year of 8,760 hours into twelve equal periods.

Average month length = 8,760 hours ÷ 12 = 730 hours

A real billing period runs 28 to 31 days, so its hours differ from this 730-hour convention, and both load factor and the adder derived from it move a little when the actual period length is used.

Load factor = Monthly kWh ÷ (Peak kW × Hours)
Default load factor = 3,000 kWh ÷ (10 kW × 730 hours)
Default load factor = 0.411, or 41.1 percent

The default business drew, on average, 41.1 percent of the power it drew during its worst interval. Load factor is not a charge on any tariff — it is a description of the relationship between usage and peak, and it is what makes the demand charge comparable with the energy rate.

Turn the demand charge into a per-kilowatt-hour adder

The demand charge can be spread across every kilowatt-hour consumed in the month. Take the demand rate and divide it by the load factor and the hours in the period.

Demand adder per kWh = Demand rate ÷ (Load factor × Hours)
Default demand adder = $15 per kW ÷ (0.411 × 730 hours)
Default demand adder = $0.0500 per kWh

That five-cent adder is not a new assumption. It is the same $150.00 demand charge, spread across 3,000 kilowatt-hours and expressed in the unit the energy rate already uses.

Demand adder = $150.00 ÷ 3,000 kWh = $0.0500 per kWh

There is an independent check available. Subtracting the stated energy rate from the effective all-in rate isolates whatever demand contributed.

Demand adder = $0.1700 per kWh − $0.1200 per kWh
Demand adder = $0.0500 per kWh

All three routes agree exactly, and that agreement is the proof of the identity: the demand rate divided by load factor and hours is the same quantity as the demand charge divided across monthly energy. It also shows why one tariff can produce sharply different effective prices, because with the rate and the period fixed, load factor alone decides how much demand cost lands on each kilowatt-hour.

Three routes, one five-cent adder Three ways in, one number out The default bill: a 10 kW peak at $15, and 3,000 kWh at 12 cents. Every route is arithmetic on those. FROM LOAD FACTOR 15 ÷ (0.411 × 730) $0.0500 per kWh FROM THE CHARGE $150 ÷ 3,000 kWh $0.0500 per kWh FROM THE ALL-IN RATE $0.1700 − $0.1200 $0.0500 per kWh They agree exactly. That is the proof. adder = demand rate ÷ (load factor × 730 h) With the rate and the month fixed, load factor alone sets the adder. 730 h is 8,760 ÷ 12. A real billing period is 28 to 31 days, so the figure moves a little either way.
The demand charge becomes a per-kilowatt-hour adder of the demand rate divided by load factor times 730 hours.
Load factorDemand adder per kWhEnergy rate per kWhAll-in rate per kWh
20.0%$0.1027$0.1200$0.2227
41.1%$0.0500$0.1200$0.1700
60.0%$0.0342$0.1200$0.1542
80.0%$0.0257$0.1200$0.1457

Every row uses the same illustrative $15-per-kilowatt demand rate and 12-cent energy rate, and only load factor changes. At 20 percent the effective price is 22.27 cents a kilowatt-hour; at 80 percent it is 14.57. The lower-load-factor business pays 52.9 percent more per kilowatt-hour on an identical rate card.

The same rate card, four different prices Same tariff. Four prices per kilowatt-hour. $15 per kW and 12 cents per kWh throughout. Only the load factor changes. LOAD FACTOR ENERGY RATE + DEMAND ADDER 20.0% $0.2227 41.1% $0.1700 the default 60.0% $0.1542 80.0% $0.1457 the floor at 100% — $0.1405 energy, 12 cents, unchanged demand adder 22.27 against 14.57 cents — 52.9 percent apart, on one rate card. Nothing about the tariff differs between the top row and the bottom. The only difference is how evenly the power was drawn across the month.
One tariff, four load factors: 22.27 cents a kilowatt-hour at 20 percent and 14.57 at 80, with no change to the rate card.

The same point can be made holding energy at exactly 3,000 kilowatt-hours in every case. A lower load factor then requires a higher peak, because the same quantity of energy is being drawn less evenly across the month.

Load factorPeak demandDemand chargeEnergy chargeTotal billDemand share
20.0%20.55 kW$308.22$360.00$668.2246.1%
41.1%10.00 kW$150.00$360.00$510.0029.4%
60.0%6.85 kW$102.74$360.00$462.7422.2%
80.0%5.14 kW$77.05$360.00$437.0517.6%

Consumption is identical in every row and the energy charge stays at $360.00 throughout. The entire spread from $437.05 to $668.22 comes from power.

The ceiling on load factor is a floor under the rate

Load factor cannot exceed 100 percent, because average power cannot be higher than the peak that average is measured against. That ceiling puts a floor under the all-in rate, and the floor is worth knowing before any effort is spent chasing it.

Best-case adder = $15 per kW ÷ (1.00 × 730 hours)
Best-case adder = $0.0205 per kWh
Best-case all-in rate = $0.1200 + $0.0205 = $0.1405 per kWh

On this illustrative tariff, demand adds at least 2.05 cents to every kilowatt-hour no matter how flat the load becomes. That is the whole reachable range: from 14.05 cents at a theoretical 100 percent, through 14.57 at 80, to 22.27 at 20 and upward from there.

A load factor of 100 percent would mean drawing exactly the same power in every hour of the month, including nights, weekends and shutdowns, which no real site does. The practical floor therefore sits above 14.05 cents, and the gap between a site's current figure and its realistic best is the size of the prize.

Why cutting energy alone can disappoint

Suppose consumption falls 20 percent, from 3,000 to 2,400 kilowatt-hours, while the 10-kilowatt peak is unchanged. The energy charge falls with the usage, and the whole $150.00 demand charge survives.

Reduced energy charge = 2,400 kWh × $0.12 per kWh = $288.00
New total bill = $150.00 + $288.00 = $438.00
Bill reduction = ($510.00 − $438.00) ÷ $510.00 × 100 = 14.1 percent

A 20.0 percent cut in usage produces a 14.1 percent cut in the bill. One component declined and the other did not move, which is all that has happened.

The effective price per kilowatt-hour actually rises, because the unchanged demand charge is now spread over fewer units. The customer buys fewer kilowatt-hours and each one carries a larger share of the demand cost.

New effective rate = $438.00 ÷ 2,400 kWh = $0.1825 per kWh
New load factor = 2,400 kWh ÷ (10 kW × 730 hours) = 32.9 percent

The all-in rate moves from $0.1700 to $0.1825 and load factor falls from 41.1 percent to 32.9. The saving is real, but the unchanged peak caps it. The energy bill estimator is the right tool where a bill follows consumption alone; once demand pricing applies, a kilowatt-hour-only view cannot explain the whole amount due.

What changes when the peak falls

Now hold consumption at 3,000 kilowatt-hours and cut the peak by 30 percent, from 10 kilowatts to 7. The energy charge does not move, and the demand charge falls with the peak.

New demand charge = 7 kW × $15 per kW = $105.00
New total bill = $105.00 + $360.00 = $465.00
Monthly saving = $510.00 − $465.00 = $45.00
Bill reduction = $45.00 ÷ $510.00 × 100 = 8.8 percent

The lower peak saves $45.00 a month, or 8.8 percent of the original bill, without saving a single kilowatt-hour. The effective rate falls to $0.1550 and the load factor rises to 58.7 percent.

New effective rate = $465.00 ÷ 3,000 kWh = $0.1550 per kWh
New load factor = 3,000 kWh ÷ (7 kW × 730 hours) = 58.7 percent

Under a flat demand rate with no other tariff rules, every kilowatt taken off the billed peak is worth that demand rate every month. On this illustrative tariff one kilowatt is $15 a month, or $180 across a year.

Monthly value per kW = 1 kW × $15 per kW = $15
Annual value per kW = $15 × 12 months = $180

This guide prices the gap between a current peak and a lower one. It does not recommend equipment, size batteries, prescribe controls, or estimate what any method of closing that gap would cost or return.

Two levers, and what a point of each is worth Two levers on one $510 bill Each moves one half and leaves the other exactly where it was. AS BILLED $150 demand $360 energy $510 USAGE −20% $150 unchanged $288 $438 bill −14.1% · all-in rate RISES 17.00 → 18.25 cents PEAK −30% $105 $360 unchanged $465 bill −8.8% · all-in rate falls to 15.50 cents · 0 kWh saved BUT COMPARE THEM POINT FOR POINT 1% off usage = $3.60 1% off peak = $1.50 ratio 2.4 — which is just $360 ÷ $150 The bigger charge always wins per point, so the levers break even at a 50% demand share. At the 29.4% default, usage is the stronger lever per point. Whether a point is equally easy to move is a site question.
Cutting usage 20 percent takes 14.1 percent off the bill; cutting the peak 30 percent takes 8.8 percent off without saving a single kilowatt-hour.

Which lever is worth more per point

The two comparisons above moved by different amounts, 20 percent against 30, so they do not settle which lever pays better. Comparing them properly means moving each by the same proportion and reading what one percentage point is worth.

One percent off usage = 30 kWh × $0.12 per kWh = $3.60
One percent off the peak = 0.1 kW × $15 per kW = $1.50
Ratio = $3.60 ÷ $1.50 = 2.4

On this bill a point of usage is worth 2.4 times a point of peak, which is the opposite of the impression the peak-shaving arithmetic tends to leave. The ratio is not a new fact either: it is just the energy charge divided by the demand charge, $360.00 over $150.00, which is the same 2.4.

That gives the demand share a job beyond description. Since the two charges are the only things being compared, the larger one always wins per point, and the levers break even exactly where the demand share reaches 50 percent. Below that, proportional cuts in usage pay more; above it, proportional cuts in peak do.

The defaults sit at a 29.4 percent demand share, so usage is the stronger lever per point there. None of which says the two are equally easy to move by a point, and that part is a question about the site rather than about the tariff.

What the two-part model leaves out

The calculator models one demand rate against one peak, plus one energy rate against monthly consumption. That clean structure is what makes load factor visible, and it does not reproduce every rule in a real tariff.

Demand ratchets can hold billed demand at a fraction of an earlier peak for several months, so lowering the current peak may not lower the billed figure straight away while a prior period still controls it. Seasonal demand rates price peaks differently across the year, and time-of-day demand windows count a peak only during specified hours, which makes the timing of demand matter as much as its size.

Coincident-peak charges can depend on the utility system's peak rather than the customer's own highest interval. Power-factor penalties can add a further charge where current and voltage are used inefficiently, and this model calculates neither.

The $15-per-kilowatt demand rate and the $0.12-per-kilowatt-hour energy rate stay illustrative inputs throughout. Real tariffs may also carry customer charges, taxes, riders, tiered energy prices, minimum bills, and their own definition of which interval sets billed demand.

Broader price context in how much does electricity cost is no substitute for the schedule that applies to the account. The tariff sheet, the billing-period dates, the demand definition and the actual interval data are the source of truth.

Questions people ask

What is load factor, and how do I calculate it?

Load factor is your average power divided by your peak power, which works out as monthly kilowatt-hours divided by the peak in kilowatts times the hours in the billing period. On the calculator defaults — 3,000 kWh against a 10 kW peak, over 8,760 ÷ 12 = 730 hours — that is 3,000 ÷ 7,300, or 41.1 percent. It means the site drew, on average, 41.1 percent of the power it drew at its worst interval. It is not a charge on any tariff, and the calculator does not compute it, because it needs a period length the engine has no input for.

Why is my effective rate higher than the rate on my tariff sheet?

Because the demand charge is money you paid for electricity that is not counted in the energy rate. Divide the whole bill by the kilowatt-hours and you get the real figure: on the defaults, $510.00 ÷ 3,000 kWh is $0.1700 per kWh against a posted energy rate of $0.1200. The five-cent gap is the demand charge spread across the energy, and it equals the demand rate divided by the load factor times the hours — 15 ÷ (0.411 × 730), which is $0.0500. The lower your load factor, the bigger that gap.

Why did cutting my usage barely change the bill?

Because energy is only one of the two charges, and cutting it leaves the other untouched. Taking usage down 20 percent, from 3,000 to 2,400 kWh, drops the energy charge from $360.00 to $288.00 while the $150.00 demand charge survives intact, so the bill falls from $510.00 to $438.00 — a 14.1 percent cut for a 20 percent reduction. The effective rate actually rises, from $0.1700 to $0.1825 per kWh, because the unchanged demand charge now spreads over fewer kilowatt-hours.

How much is one kilowatt off the peak worth?

Under a flat demand rate with no other tariff rules, exactly the demand rate, every month. On the illustrative $15 per kW used here, one kilowatt is $15 a month or $180 a year, and it arrives whether or not you use any less electricity. Taking the default peak from 10 kW to 7 kW cuts the demand charge from $150.00 to $105.00 and the bill from $510.00 to $465.00 — a $45.00 saving, 8.8 percent of the bill, with not a single kilowatt-hour saved. Use your own tariff figure rather than the $15.

What is the lowest my all-in rate can go?

There is a floor, because load factor cannot exceed 100 percent — average power cannot be higher than the peak it is measured against. At a theoretical 100 percent, the adder on the illustrative tariff here is 15 ÷ (1.00 × 730), or $0.0205 per kWh, giving an all-in rate of $0.1405. That is the best the tariff can do. A 100 percent load factor would mean drawing identical power in every hour of the month, including nights and shutdowns, so the practical floor sits above it.

Does a lower load factor always mean a higher bill?

It always means a higher price per kilowatt-hour on the same tariff, which is not the same as a higher bill. Holding energy at 3,000 kWh, a 20 percent load factor needs a 20.55 kW peak and totals $668.22, while an 80 percent load factor needs only 5.14 kW and totals $437.05 — same energy, same rates, a $231 spread. But a site that uses far less electricity can have a poor load factor and still pay less overall, because the energy charge fell too. The rate per kilowatt-hour is what load factor governs.

Do residential customers pay demand charges?

Not as a rule. The two-part structure of a demand charge plus an energy charge is normally associated with commercial and industrial accounts, which is why the concept surprises people the first time they take on a commercial meter. Some utilities have introduced residential demand rates in particular service territories, so it is not unheard of, but a residential bill should not be assumed to carry one. The tariff schedule for the account settles it.

What does this two-part model leave out?

A good deal, and all of it can change what you owe. Demand ratchets can hold billed demand at a fraction of an earlier peak for months, so a lower peak now may not lower the charge yet. Seasonal demand rates price peaks differently across the year, and time-of-day windows count a peak only in specified hours. Coincident-peak charges can key off the utility system’s peak rather than yours, and power-factor penalties are a separate charge entirely. Customer charges, taxes, riders, tiered rates and minimum bills sit outside it too.